Bacterial vaginosis (BV) is a condition that affects millions of women, produces malodorous vaginal discharge, and is linked to several serious health conditions including preterm labor and HIV infection. The cause of BV is unclear, though changes in vaginal flora are noted and patients may respond to antibiotic treatment. No single cultivated bacterium is specifically associated with BV. Bacteria in complex microbial communities can be identified without cultivation by characterizing ribosomal DNA sequences. This approach offers the advantage of detecting fastidious or cultivation-resistant microbes. The short-term objective of this project is to use cultivation-independent molecular methods to characterize the vaginal flora of women with BV and healthy controls. The long-term objectives are to understand the bacterial ecology of the vaginal niche and determine the role uncultivated bacteria play in BV. Hypothesis: novel communities of bacteria, including many uncultivated species, are found in subjects with BV. To meet these objectives, the following aims will be pursued: Aim 1:Identify the community of bacteria associated with BV. Broad range PCR will be used to amplify bacterial 16SrRNA genes present in vaginal fluid. Sequence analysis of the amplified 16SrDNA will be used to identify the bacterial species found in subjects with BV and healthycontrols. Aim 2: Usebacterium specific PCR and fluorescence in situ hybridization (FISH) assays to detect and quantify each vaginal bacterium. Bacterium-specificPCR and FISH assays are more sensitive than broad range PCR for detecting infrequent bacterial species in complex communities. These methods will be applied to vaginal fluid samples from additional subjects with BV and healthy controls. FISH paired with fluorescence microscopy will be used to count bacteria in vaginal fluid smears, thus producing an independent measure of bacterial representation to complement PCR methods. Aim3:Determine if there is an extra-vaginal niche for BVassociated bacteria. The gastrointestinal tract hosts numerous anaerobic bacteria and may be a reservoir for the bacteria linked to BV. Swabs ofextra-vaginal mucosal surfaces will be obtained from women with and without BV. Bacteria on these swabs will be detectedusing bacterium-specific PCRs. We hypothesize that BV-associated bacteria will be common colonizers of thehuman gastrointestinal tract. Knowledge gained from this project will help identify the bacterial communities found in subjects with BV, will generate opportunities to advance the diagnosis of BV through new PCR and FISH assays, and may help in the prevention of BV by identifying extra-vaginal reservoirs of infection.